When speaking to an adult, a small child may be motivated to speak clearly, so as to avoid miscommunication. Given the capabilities of modern ASR systems to accurately recognize a wide range of speech, including words that are incorrectly pronounced, a small child may be less motivated to speak clearly when speaking a command into a computing device.